Automatic Identification of MALDI-TOF MS Database Using Classical <i>Bordetella</i> Species Isolates

Liu, Yamin; Cui, Junwen; Qie, Chunhua; Jiang, Bei; Li, Ying; Zhao, Xiaoyun

doi:https://doi.org/10.1155/2022/1679951

Computational and Mathematical Methods in Medicine

On this page

Abstract Introduction Materials and Methods Results Discussion Conclusion Data Availability Conflicts of Interest Authors’ Contributions Acknowledgments References Copyright Related Articles

Special Issue

Health Informatics: Computer Algorithms in Operational Navigation and Medical Data Mining

View this Special Issue

Research Article | Open Access

Volume 2022 | Article ID 1679951 | https://doi.org/10.1155/2022/1679951

Automatic Identification of MALDI-TOF MS Database Using Classical Bordetella Species Isolates

Yamin Liu,¹Junwen Cui,²Chunhua Qie,¹Bei Jiang,³Ying Li,⁴and Xiaoyun Zhao^5,6

Academic Editor: Xue Fei Deng

Received31 Mar 2022

Revised24 Apr 2022

Accepted24 May 2022

Published15 Jun 2022

Abstract

Objective. To evaluate and expand the automatic identification and clustering of clinical Bordetella species by MALDI-TOF MS. Methods. Twenty-eight field isolated strains, identified by whole-gene sequencing analysis, were analyzed by MALDI-TOF MS, and the spectra obtained were used to replenish the internal database of the manufacturer. To evaluate and expand the robustness of the database, MALDI-TOF MS identified 91 clinical isolates (except those used for implementation). A distance tree based on mass spectrometry data is constructed to confirm similarity and clusters of each clinical Bordetella species by using the MALDI Biotyper 3.1 software. Results. In this research, when we used the implemented Bruker Daltonics database in our laboratory, 91 clinical isolates were identified at the genus level (100%) and 93.4% were identified at the species level (85/91). We performed proteomics analysis and divided these 91 isolates into cluster I (2.2%) and cluster II (97.8%). The largest group is cluster II ( isolates), which has been divided into two subclusters. Trees created by analyzing the protein mass spectra of the three species of the clinical isolates reflected their classification. Conclusion. MALDI-TOF MS may present an attractive alternative to automatically confirm and cluster the fastidious bacteria difficult to culture. Extension of identification of the MALDI-TOF MS database is viably fast, more efficient, and alternative to conventional methods in confirming the classical Bordetella species. This strategy could promote the epidemiological and taxonomic research of this important pathogen.

1. Introduction

Pertussis disease, also known as whooping cough, is a vaccine-preventable infectious disease caused by the bacterium called Bordetella pertussis (Bp). Infants too young to be vaccinated or not fully vaccinated are the most severely infected groups [1, 2]. Despite the wide coverage of vaccination, the resurgence of pertussis has been observed in many countries around the world [3, 4]. There are additional members of the Bordetella family, such as Bordetella parapertussis (Bpp) and Bordetella bronchiseptica (Bb) (usually known as the classical Bordetella species), which are also related to respiratory tract infections in humans or animals.

To implement effective prevention and control measures for Bordetella pertussis, we should be able to confirm this pathogen precisely. Bacterial culture has been considered to be the gold standard for detection of Bordetella pertussis. In the past years, the identification of Bp has mainly depended on a biochemical test, slide agglutination, and polymerase chain reaction (PCR) [5]. But these techniques have many shortcomings and require certain professional knowledge to accurately confirm and classify [6].

Recently, MALDI-TOF MS has become a breakthrough method for rapid, routine identification of microorganisms with excellent cost-effectiveness, high reproducibility, and reliability. The principle of this technology is to extract unique and representative spectral signals from microorganisms and establish the corresponding recognition relationship by matching the signal pattern with the pattern of the reference strain in the database. It is a rapid and reliable microbial identification and clustering technology [7, 8]. However, MALDI-TOF MS for identification and cluster analysis on Bordetella species has rarely been reported. The commercialized Bruker Biotyper database (DB-5989MSP) available contained Bordetella species with small amount of bacterial spectrum information, but the spectrum database of other brands of mass spectrometers has not included Bordetella species spectrum information. In order to make it possible to confirm the isolated classic Bordetella species more accurately in our region, we used clinical strains to evaluate and expand the capability of the MALDI Biotyper system to confirm, cluster, and analyze classical Bordetella species.

2. Materials and Methods

2.1. Bacterial Strains

A total of 119 Bordetella species isolates from infants of 0-6 months with suspected whooping cough were collected from 2018 to 2019 in the infectious center of Tianjin Second People’s Hospital, China. The isolated strains were mainly from nasopharyngeal swabs or nasopharyngeal aspirates. Bacterial strains were cultured on charcoal agar (Oxoid Ltd., England) with the addition of 10% sheep blood and 40 μg/ml cephalexin (Bordetella Selective Supplement; Oxoid Ltd.) (BORDC medium) at 37°C with 5% CO₂ for 3-7 days and inspected regularly, as previously reported [9]. Presumptive Bordetella colonies with a grey morphology were measured by MALDI-TOF MS. Twenty-eight field isolated strains were identified by whole-gene sequencing analysis performed by MALDI-TOF MS, and these spectra were utilized to replenish the manufacturer internal data (Bruker Daltonik GmbH). In order to evaluate and expand the robustness of the database, MALDI-TOF MS identified 91 clinical isolates (except those used for implementation).

2.2. Sample Preparation for MALDI Biotyper Analysis

Proteides were pretreated using the procedure of ethanol/formic acid extraction as previously described [10]. Suspended bacterial colonies in 300 μl distilled water and 900 μl of ethyl alcohol were vortexed for 30 seconds and then centrifuged at for 2 minutes. Remove the supernatant, then centrifuge 2 times, remove excess ethanol, then dry at common temperature, and resuspend the precipitate in 50 μl of 70% formic acid aqueous solution and 50 μl in acetonitrile solution. Mix samples thoroughly at for 2 minutes, and spot 1 μl supernatant onto a 96-target polished steel plate (Bruker Daltonik). Dry at room temperature, and 1 μl of HCCA (alpha-cyano-4-hydroxycinnamic acid) was dropped on the sample and waited to crystallize.

2.3. MALDI-TOF MS Database Expansion for Bordetella Species

The commercialized Bruker Biotyper database (DB-5989MSP) available already contained 10 spectra for Bp, 11 spectra for Bpp, and 9 spectra for Bb. This database also contained spectra for the other species of Bordetella, except Bordetella ansorpii. In order to make it possible to confirm the isolated classic Bordetella species more accurately in our region, we expanded the commercialize Bruker Biotyper library in our experience. Each of the 28 strains was analyzed and supplemented in the Bruker Biotyper library (DB-5989MSP) confirmed by MALDI-TOF MS and Whole-Genome Sequencing (WGS). BTS was used as an internal calibration procedure. Spot each sample at 8 locations of the target and about 24 original spectra of each strain. The spectra obtained were analyzed by using the flexAnalysis software (version 3.0, Bruker Daltonics) for “smoothing” and “baseline,” and the spectrum with intensity less than 10⁴ unitary units and singlet spectra with different peaks were selected. As described previously [11], these selected spectra were removed by using the automation function on the MALDI Biotyper software (Biotyper MSP Creation Standard Method, Bruker Daltonics), and the rest was used to reckon the reference in the main spectral profile (MSP).

For the sake of obtaining typical results, at least 20 spectra were applied to set up a singlet MSP proponent by Bruker Daltonik GmbH (Bremen, Germany). Although the automation function on MALDI Biotyper software was utilized, these remaining required spectra were used to reckon the reference in MSP. The MSP spectra were used for MALDI Biotyper database expansion and implementation.

2.4. MALDI Biotyper Analysis

The MALDI Biotyper RTC software version 3.1 (Bruker Daltonik) was used for bacterial identification. Kations were pretreated at accelerating voltage of 20 kV, and spectra were analyzed in the positive linear mode at a mass charge ratio () of 2,000 to 20,000 Da. Adjust the laser intensity of the sample to be slightly above the ionization threshold and desorption thresholds. Each optical spectrum represents a single proteide spectrum of a bacterial strain. After automatic acquisition, the spectrum of the test bacteria, acquired through the MALDI Biotyper RTC, is transformed into a peak list. The bacterial test standard (BTS) was calibrated by using a Bruker Daltonics instrument. And the spectra were analyzed by using the Bruker Biotyper library (DB-5989MSP) and the expanded in-house MALDI-TOF MS database. 10 best matches and the corresponding matching score were shown in the result report for each sample. According to the matching MSP and the corresponding final logarithmic score, the identification results are displayed in the score table.

According to the Bruker’s proprietary algorithm, this peak list is compared to the reference peaks of organism in the reference database, and a score value range from zero to three is calculated. The high score means a high degree of similarity compared with a given organism in the reference database. Briefly, 0 to 1.699 indicates that there is no reliable identification, 1.700 to 1.999 indicates a possible genus-level identification, 2.000 to 2.299 indicates a safe genus-level identification and a possible species-level identification, and 2.300 to 3.000 indicates a highly possible species-level identification [12].

2.5. Principle Component Analysis (PCA) of the Analyzed Genomic Strains

To compare each strain, the spectra of the 91 examined strains were loaded into the program ClinProTools for automatic recalibration. Generate classification models. Here, specific software algorithms were used, including Quick Classifier (QC)/Different Average, Supervised Neural Network (SNN), and Genetic Algorithm (GN). According to the selected algorithm, these algorithms propose the list of identification peaks of the analytical spectrum. The statistical test of the dataset was carried out on the basis of principal component analysis (PCA), and the results were exhibited in the three-dimensional score map automatically created by that software.

2.6. Cluster Analysis

For MALDI-TOF MS cluster analysis, 91 isolated straints reference spectra, following be called MSPs (Main Spectral Profiles), independently created. Each of the extracted samples of the prepared representative strain was dropped on 8 points on the polished steel plate and then tested three times by averaging data from 240 laser shots using MALDI Biotyper software to generate 24 qualified mass spectra, which were subsequently composed of one MSP. MSPs of the strains were used to cluster the classical Bordetella species. Based on the values obtained from the MSPs of each strain, a tree was generated, allowing visualization of the similarities among all spectrum profiles. Use the comprehensive statistical tool MATLAB 7.1 of the Biotyper 3.1 software toolbox to process hierarchical cluster analysis and the default correlation function.

3. Results

3.1. Accuracy of Identification by MALDI-TOF MS

All 91 strains were subjected to analysis by MALDI Biotyper RTC software version 3.1, and acquired spectra were aligned to the Bruker Biotyper library (DB-5989MSP) and the implemented database. The results of the log score obtained by MALDI-TOF MS identification are shown in Table 1. According to the results of MALDI-TOF MS, classify the 91 strains as 89 Bp, 1 Bpp, and 1 Bb at the species level. When using the implemented Bruker Daltonics database in our laboratory, all isolated strains were confirmed at the genus level (100%) and 93.4% were confirmed at the species level (85/91). Bp strains were 89 cases. Among them, there were 60.7% (54/89) strains with the reliability score of 2.0-2.229 (accurately identified genus) before the expansion of the Bp database. There were 37.1% (33/89) strains with a score of 1.700-1.999 and only 2.2% (2/89) strains with a (accurately identified species). After the expansion of the database, there were 93.3% (83/89) strains with a reliability score of >2.300 and 67.4% (6/89) strains with a score of 2.000-2.229 and no strains with a score of <2.000.

3.2. Principle Component Analysis (PCA) of the Analyzed Genomic Strains

The PCA software in ClinProTools was used for the analysis database in visualizing the homogeny and neterogeny of these proteide spectrum. PCA can debase the diverse complex datasets based on different statistical tests. The reduced datasets, which are called PCs (principle components), can be exhibited in a plot diagram. A total of 91 individual protein spectra of the clinical Bordetella species are exhibited in a three-dimensional PCA in Figure 1. Each plot diagram means an exhibited proteide spectrum. These colors stand for the counted cluster members, where each point delegates a measured protein mass spectrum profile of each sample.

3.3. Cluster Analysis

Create an MSP tree (Figure 2) to visually compare the relationships between each mass spectrum got from these 91 isolates. For a preset bacterial strain, use the rigorous data processing algorithms on MALDI Biotyper 3.1 software for creating the spectrum profile for each sample. The mass spectrum patterns were used to study the diverse technicality for confirming 91 different isolated strains. In the MSP tree view, the relative distance among isolates was exhibited as an arbitrary unit. Zero means complete similarity, and 1000 means complete dissimilarity. The willful distance level of 500 was selected for cluster evaluation of isolated strains. Our cluster analysis of all MSPs divided these Bordetella isolates into two clusters: cluster I, which contained 2 (2.2%) isolates, and cluster II, which contained 89 (97.8%) isolates. Cluster II, as the largest group, has been divided into two subclusters. Figure 3 shows that all peaks are exhibited for the 89 Bp (black), 1 Bpp (red), and 1 Bb (green), and ion intensities, peak positions, and peak frequencies derived from the 91 isolates, representative strain of the two clusters, are the function of the values. Total peaks for the 91 isolates were produced by flexAnalysis 3.3 software. Representative strains of mass spectrometric profiles of Bp and Bpp isolates are shown in Figure 4.

4. Discussion

Bordetella pertussis, as an exclusive human pathogen, can cause pertussis disease, a contagious infection in the human respiratory tract [13]. Pertussis continues to be an important cause of morbidity worldwide and of mortality in infants, especially under 6 months [14, 15]. The symptoms in clinic vary with age, former infection with B. pertussis, and vaccinated status. In those with incomplete immunity, pertussis can develop in atypical clinical forms and is difficult to be diagnosed [16]). The clinical manifestations in newborns may be severe. Most infants have typical spasmodic paroxysmal cough that lasts for more than two months [17]. Pertussis in children under 1 year of age is particularly difficult to diagnose in winter because of other pathogens such as influenza or respiratory syncytial virus (RSV) epidemics [18, 19].

In clinic practice, diagnosis is often given without any microbiological basis leading to a possibility that cannot initiate treatment early and prevent complications primitively. Confirming Bordetella pertussis in the laboratory is difficult and contributes to underreporting of the disease [20]. The operation method of the biochemical test is cumbersome and time-consuming, and slide agglutination requires a special antigen reagent, and both the false positive and false negative rates are high, also needing a professional operator [6]. There is an urgent need for a reliable device for rapid diagnosis of pertussis. By MALDI-TOF MS entering the clinical laboratory, the identification of this nutrient-critical and slow-growing bacterium has been improved [21]. In the past few years, MALDI-TOF MS has gradually been extensively used for quickly confirming and typing a variety of bacteria and fungi in the clinical microbiological laboratory [22–24]. This MALDI-TOF MS identification is mainly based on the differences of whole proteins, especially ribosomal proteins, which are automatically matched with the database. Furthermore, the technology will be automated, has high throughput, and is suitable for a wide range of common and esoteric bacteria and fungi [25, 26]. Due to the large regional differences of Bp strains, the existing mass spectrometry library cannot meet the needs of accurate identification of local epidemic strains.

In this work, we analyzed and supplemented 28 strains in the Bruker Biotyper library (DB-5989MSP), which had been confirmed by Whole-Genome Sequencing (WGS). Then, we evaluated the performance of MALDI-TOF MS to rapidly detect the Bordetella species, using 91 previously characterized isolates, in order to consider it as an alternative tool for confirming the fastidious bacteria within minutes instead of hours and days. However, when using the implemented Bruker Daltonics database in our laboratory, all of isolates were confirmed correctly at the genus level (100%) and 93.4% (85/91) were confirmed at the species level. Bp species was confirmed correctly at the genus level (100%), and 93.3% (83/89) was confirmed at the species level. These results show that the accurate identification of bacteria depends on the quality of the database. Thence, the Biotyper database needs to be expanded by additional more spectra to create one representative database, to promote the matching score and achieve highly precise species-level identification. The implementation of the database reinforced the scores of identification of the classical Bordetella species. This will be very important for future applications in clinics.

A further main objective of our study was to assess the capability MALDI-TOF for the clustering of the Bordetella species. MALDI-TOF has been reported to be successfully applied for cluster analysis of many pathogens [27, 28]. Based on the data obtained from the paired comparisons of different spectra, a tree view was generated, which allows visualization of the similarity among all the spectral profiles. Our cluster analysis of all major spectra classified these Bordetella species to two clusters: cluster I, which contained 2 isolates (2.2%), and cluster II, which contained 89 isolates (97.8%). Except for the bacterial culture time, only 30 minutes were necessary to carry out the entire identification process, starting from the picking of the single colony and ending in analysis by MALDI-TOF MS. The MSP dendrogram is considered a reliable tool for illustrating the capability of MALDI-TOF MS to visualize the degree of similarities and differences between species when more isolates are considered [29, 30]. The manifestation of MALDI-TOF for cluster analysis was evaluated as one extended and complementary typing method, especially for the screening purpose. We expanded the sensitive culture method by screening with MALDI-TOF MS for the identification and clustering of the classical Bordetella species. Automatic identification and clinical analysis may be achieved by deep learning or predictive control techniques [31] and may be useful in the performance evaluation of analytical procedures in the future.

5. Conclusion

In summary, our study strongly indicates that using classical Bordetella species isolates can expand automatic identification of the MALDI-TOF MS database. Extension of identification of the MALDI-TOF MS database is viably fast, more efficient, and alternative to conventional methods and may aid in the surveillance of regional strains of the classical Bordetella species. It may present an attractive alternative to confirm and cluster the fastidious bacteria difficult to culture. This strategy could automatically facilitate the epidemiological and taxonomic research of the important pathogen.

Data Availability

Data are available on request from the authors due to privacy/ethical restrictions.

Conflicts of Interest

The authors declare no conflict of interest for this paper.

Authors’ Contributions

Yamin Liu and Junwen Cui are co-first authors.

Acknowledgments

This study was supported by the Tianjin Health Commission Tianjin Administration of Traditional Chinese Medicine Integrated Traditional Chinese and Western Medicine Scientifc Research Project (Nos. 2019131 and 2021138).

References

V. Bouchez, J. Guglielmini, M. Dazas et al., “Genomic sequencing of Bordetella pertussis for epidemiology and global surveillance of whooping cough,” Emerging Infectious Diseases, vol. 24, no. 6, pp. 988–994, 2018.
View at: Publisher Site | Google Scholar
R. S. Torres, T. Z. Santos, R. A. Torres et al., “Ressurgimento da coqueluche na era vacinal: aspectos clinicos, epidemiologicos e moleculares,” Jornal de Pediatria, vol. 91, no. 4, pp. 333–338, 2015.
View at: Publisher Site | Google Scholar
K. L. Sealey, T. Belcher, and A. Preston, “Bordetella pertussis epidemiology and evolution in the light of pertussis resurgence,” Infection, Genetics and Evolution, vol. 40, pp. 136–143, 2016.
View at: Publisher Site | Google Scholar
E. Chiappini, A. Stival, L. Galli, and M. de Martino, “Pertussis re-emergence in the post-vaccination era,” BMC Infectious Diseases, vol. 13, no. 1, p. 151, 2013.
View at: Publisher Site | Google Scholar
Q. Zhang, Z. Yin, Y. Li et al., “Prevalence of asymptomatic Bordetella pertussis and Bordetella parapertussis infections among school children in China as determined by pooled real-time PCR: a cross-sectional study,” Scandinavian Journal of Infectious Diseases, vol. 46, no. 4, pp. 280–287, 2014.
View at: Publisher Site | Google Scholar
V. S. Nikbin, F. Shahcheraghi, M. N. Lotfi, S. M. Zahraei, and M. Parzadeh, “Comparison of culture and real-time PCR for detection of Bordetella pertussis isolated from patients in Iran,” Iranian Journal of Microbiology, vol. 5, no. 3, pp. 209–214, 2013.
View at: Google Scholar
M. Ricchi, A. Mazzarelli, A. Piscini et al., “Exploring MALDI-TOF MS approach for a rapid identification of Mycobacterium avium ssp. paratuberculosis field isolates,” Journal of Applied Microbiology, vol. 122, no. 3, pp. 568–577, 2017.
View at: Publisher Site | Google Scholar
S. W. Kim, S. W. Nho, S. P. Im et al., “Rapid MALDI biotyper-based identification and cluster analysis of Streptococcus iniae,” Journal of Microbiology, vol. 55, no. 4, pp. 260–266, 2017.
View at: Publisher Site | Google Scholar
D. Van den Bossche, A. De Bel, D. De Smet et al., “Prevalence of Bordetella holmesii and Bordetella bronchiseptica in respiratory tract samples from Belgian patients with pertussis-like symptoms by sensitive culture method and mass spectrometry,” Acta Clinica Belgica, vol. 68, no. 5, pp. 341–348, 2013.
View at: Publisher Site | Google Scholar
V. G. Rocco, J. Intra, C. Sarto et al., “Rapid identification of carbapenemase-producing Klebsiella pneumoniae strains by matrix-assisted laser desorption/ionization-time of flight using Vitek® mass spectrometry system,” The Eurasian Journal of Medicine, vol. 51, no. 3, pp. 209–213, 2019.
View at: Publisher Site | Google Scholar
T. Bartosch, T. Heydel, S. Uhrlaß et al., “MALDI-TOF MS analysis of bovine and zoonotic Trichophyton verrucosum isolates reveals a distinct peak and cluster formation of a subgroup with Trichophyton benhamiae,” Medical Mycology, vol. 56, no. 5, pp. 602–609, 2018.
View at: Publisher Site | Google Scholar
A. Cherkaoui, J. Hibbs, S. Emonet et al., “Comparison of two matrix-assisted laser desorption ionization-time of flight mass spectrometry methods with conventional phenotypic identification for routine identification of bacteria to the species level,” Journal of Clinical Microbiology, vol. 48, no. 4, pp. 1169–1175, 2010.
View at: Publisher Site | Google Scholar
M. B. Machado and S. D. Passos, “Severe pertussis in childhood: update and controversy - systematic review,” Revista Paulista de Pediatria, vol. 37, no. 3, pp. 351–362, 2019.
View at: Publisher Site | Google Scholar
S. L. Boulet, A. T. Chamberlain, H. H. Biswas, and D. J. Jamieson, “Trends in infant pertussis hospitalizations in the United States, 2009-2017,” JAMA, vol. 322, no. 21, pp. 2134–2136, 2019.
View at: Publisher Site | Google Scholar
K. S. Marshall, H. E. Quinn, A. J. Pillsbury, J. E. Maguire, R. M. Lucas, and A. Dey, “Australian vaccine preventable disease epidemiological review series: pertussis, 2013–2018,” Communicable Diseases Intelligence, vol. 46, 2022.
View at: Publisher Site | Google Scholar
J. Pan, Y. Wang, L. Cao et al., “Impact of immunization programs on 11 childhood vaccine-preventable diseases in China: 1950-2018,” The Innovation, vol. 2, no. 2, article 100113, 2021.
View at: Publisher Site | Google Scholar
P. E. Kilgore, A. M. Salim, M. J. Zervos, and H. J. Schmitt, “Pertussis: microbiology, disease, treatment, and prevention,” Clinical Microbiology Reviews, vol. 29, no. 3, pp. 449–486, 2016.
View at: Publisher Site | Google Scholar
A. Frassanito, R. Nenna, A. Nicolai et al., “Infants hospitalized for Bordetella pertussis infection commonly have respiratory viral coinfections,” BMC Infectious Diseases, vol. 17, no. 1, article 492, p. 492, 2017.
View at: Publisher Site | Google Scholar
P. Stefanelli, G. Buttinelli, P. Vacca et al., “Severe pertussis infection in infants less than 6 months of age: clinical manifestations and molecular characterization,” Human Vaccines & Immunotherapeutics, vol. 13, no. 5, pp. 1073–1077, 2017.
View at: Publisher Site | Google Scholar
A. Zouari, H. Smaoui, D. Brun et al., “Prevalence of Bordetella pertussis and Bordetella parapertussis infections in Tunisian hospitalized infants: results of a 4-year prospective study,” Diagnostic Microbiology and Infectious Disease, vol. 72, no. 4, pp. 303–317, 2012.
View at: Publisher Site | Google Scholar
S. Biswas and J. M. Rolain, “Use of MALDI-TOF mass spectrometry for identification of bacteria that are difficult to culture,” Journal of Microbiological Methods, vol. 92, no. 1, pp. 14–24, 2013.
View at: Publisher Site | Google Scholar
X.-F. Chen, X. Hou, M. Xiao et al., “Matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) analysis for the identification of pathogenic microorganisms: a review,” Microorganisms, vol. 9, no. 7, article 1536, 2021.
View at: Publisher Site | Google Scholar
L. P. M. Siqueira, V. M. F. Gimenes, R. S. de Freitas et al., “Evaluation of Vitek MS for differentiation of Cryptococcus neoformans and Cryptococcus gattii genotypes,” Journal of Clinical Microbiology, vol. 57, no. 1, article e01282, 2019.
View at: Publisher Site | Google Scholar
G. Bacanelli, L. C. Olarte, M. R. Silva et al., “Matrix assisted laser desorption ionization-time-of-flight mass spectrometry identification of Mycobacterium bovis in Bovinae,” The Journal of Veterinary Medical Science, vol. 81, no. 10, pp. 1400–1408, 2019.
View at: Publisher Site | Google Scholar
D. Huschek and K. Witzel, “Rapid dereplication of microbial isolates using matrix-assisted laser desorption ionization time-of-flight mass spectrometry: a mini-review,” Journal of Advanced Research, vol. 19, no. 19, pp. 99–104, 2019.
View at: Publisher Site | Google Scholar
A. Van Belkum, M. Welker, D. Pincus, J. P. Charrier, and V. Girard, “Matrix-assisted laser desorption ionization time-of-flight mass spectrometry in clinical microbiology: what are the current issues?” Annals of Laboratory Medicine, vol. 37, no. 6, pp. 475–483, 2017.
View at: Publisher Site | Google Scholar
J. Li, W. W. Hu, F. L. Zhang, M. Li, C. L. Rao, and W. P. Lu, “Evaluation of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry for identifying Burkholderia pseudomallei and Burkholderia thailandensis isolates,” European Journal of Clinical Microbiology & Infectious Diseases, vol. 38, no. 1, pp. 191–196, 2019.
View at: Publisher Site | Google Scholar
É. F. Wigmann, J. Behr, R. F. Vogel, and L. Niessen, “MALDI-TOF MS fingerprinting for identification and differentiation of species within the Fusarium fujikuroi species complex,” Applied Microbiology and Biotechnology, vol. 103, no. 13, pp. 5323–5337, 2019.
View at: Publisher Site | Google Scholar
J. J. Christensen, R. Dargis, M. Hammer et al., “Matrix-assisted laser desorption ionization-time of flight mass spectrometry analysis of Gram-positive, catalase-negative cocci not belonging to the Streptococcus or Enterococcus genus and benefits of database Extension,” Journal of Clinical Microbiology, vol. 50, no. 5, pp. 1787–1791, 2012.
View at: Publisher Site | Google Scholar
P. Honnavar, A. K. Ghosh, S. Paul et al., “Identification of Malassezia species by MALDI-TOF MS after expansion of database,” Diagnostic Microbiology and Infectious Disease, vol. 92, no. 2, pp. 118–123, 2018.
View at: Publisher Site | Google Scholar
K. K. L. Wong, G. Fortino, and D. Abbott, “Deep learning-based cardiovascular image diagnosis: a promising challenge,” Future Generation Computer Systems, vol. 110, pp. 802–811, 2020.
View at: Publisher Site | Google Scholar

Copyright

Copyright © 2022 Yamin Liu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

304

Downloads

437

Citations